Skip to content

Update en install index #7285

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 4 commits into from
Apr 24, 2025
Merged

Update en install index #7285

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
cac878a
Update en install index
tianshuo78520a Apr 23, 2025
36e0562
Update
tianshuo78520a Apr 23, 2025
f541e7b
Update
tianshuo78520a Apr 23, 2025
b9c1c42
Update
tianshuo78520a Apr 23, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
179 changes: 48 additions & 131 deletions docs/install/index_en.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,8 +9,7 @@
Important updates
----------------------

* Add support for python3.12, and no longer supports python3.7
* Add support for CUDA 12.0, and no longer supports CUDA 10.2
* Paddle supports user installation without depending on CUDA and cuDNN, and automatically handles the installation of CUDA and cuDNN.


------------------------
Expand All @@ -23,10 +22,10 @@ The manuals will guide you to build and install PaddlePaddle on your 64-bit desk
1. Operating system requirements:
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

* Windows 7 / 8 / 10, Pro/Enterprise
* Ubuntu 18.04 / 20.04
* CentOS 7
* macOS 10.x/11.x/12.x/13.x/14.x
* Windows 7 / 8 / 10 / 11, Pro/Enterprise
* Ubuntu 20.04 / 22.04
* almalinux 8
* macOS 12.x/13.x/14.x/15.x
* 64-bit operating system is required

2. Processor requirements:
Expand All @@ -38,180 +37,99 @@ The manuals will guide you to build and install PaddlePaddle on your 64-bit desk
3. Version requirements of python and pip:
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

* Python requires version 3.8/3.9/3.10/3.11/3.12
* Python requires version 3.8/3.9/3.10/3.11/3.12/3.13
* Python needs pip, and pip requires version 20.2.2 or above
* Python and pip requires 64-bit

4. PaddlePaddle's support for GPU:
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
**First Installation Method: Using pip for installation**
You can choose any of the four methods: "Using pip for installation", "Using conda for installation", "Using docker for installation", "Compiling from source code for installation".

* Currently, **PaddlePaddle** supports **CUDA** driver of **NVIDIA** graphics card and **ROCm** driver of **AMD** card.
* You need to install `cuDNN <https://docs.nvidia.com/deeplearning/sdk/cudnn-install/>`_ , and version 7.6+ is required(For CUDA11)
* If you need GPU multi-card mode, you need to install `NCCL 2 <https://developer.nvidia.com/nccl/>`_
This section will introduce the installation method using pip.

* Only Ubuntu/CentOS support NCCL 2
* You need to install `CUDA <https://docs.nvidia.com/cuda/cuda-installation-guide-windows/>`_ , depending on your system, there are different requirements for CUDA version:
1.You need to ensure that your operating system meets the requirements listed above.

* Windows install GPU version
2.You need to ensure that your processor meets the requirements listed above.

* Windows 7 / 8 / 10 support CUDA 11.0/11.2/11.6/11.8/12.0 single-card mode
* don't support install using **nvidia-docker**
* Ubuntu install GPU version
3.Ensure that the Python where you need to install PaddlePaddle is in your expected location, as your computer may have multiple Pythons.

* Ubuntu 18.04 supports CUDA (11.0 - 12.0)
* Ubuntu 20.04 supports CUDA (11.0 - 12.0)
* If you install using **nvidia-docker** , it supports CUDA 11.2/11.7/12.0
* CentOS install GPU version
Use the following command to output the Python path, depending on your environment you may need to replace all the python in the command line in the instructions with the specific Python path.

* If you install using native **pip** :
In the Windows environment, the command to output the Python path is:

* CentOS 7 supports CUDA (11.0 - 12.0)
* If you compile and install using native source code:
::

* CentOS 7 supports CUDA (11.0 - 12.0)
* If you install using **nvidia-docker** , CentOS 7 supports CUDA 11.2/11.7/12.0
* macOS isn't supported: PaddlePaddle has no GPU support in Mac OS platform
where python

Please make sure your environment meets the above conditions. If you have other requirements, please refer to `Appendix <Tables_en.html#ciwhls-release>`_ .
In the macOS/Linux environment, the command to output the Python path is:

5. PaddlePaddle's support for NCCL:
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
::

* Support for Windows
which python

* not support NCCL
* Support for Ubuntu
4.Check the Python version

* Ubuntu 18.04:
Use the following command to confirm it is 3.8/3.9/3.10/3.11/3.12/3.13

* support NCCL v2.4.7 / v2.16.5 under CUDA11
* Ubuntu 20.04:

* support v2.4.7 / 2.16.5 under CUDA11
* Support for CentOS

* CentOS 7:

* support NCCL v2.4.7-v2.16.5 under CUDA11
* Support for macOS

* not support NCCL


The first way to install: use pip to install
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

You can choose any of the four ways to install: "use pip to install", "use Conda to install", "use Docker to install", "compiling from the source code"

This section describes how to use pip to install.

1. You need to confirm that your operating system meets the requirements listed above

2. You need to confirm that your processor meets the requirements listed above

3. Confirm that the Python where you need to install PaddlePaddle is your expected location, because your computer may have multiple Python

Use the following command to output Python path. Depending on your environment, you may need to replace Python in all command lines in the description with specific Python path

In the Windows environment, the command to output Python path is:

::

where python

In the macOS/Linux environment, the command to output Python path is:

::

which python


4. Check the version of Python

Confirm the Python is 3.8/3.9/3.10/3.11/3.12 using command
::

python --version

5. Check the version of pip and confirm it is 20.2.2 or above
5.Check the pip version, confirm it is 20.2.2+

::

python -m ensurepip
python -m pip --version


6. Confirm that Python and pip is 64 bit，and the processor architecture is x86_64(or called x64、Intel 64、AMD64)architecture. Currently. The first line below outputs "64bit", and the second line outputs "x86_64", "x64" or "AMD64" :
6.Confirm that Python and pip are 64 bit, and the processor architecture is x86_64 (also known as x64, Intel 64, AMD64). The first line of the following output is "64bit", and the second line output is "x86_64", "x64" or "AMD64":

::

python -c "import platform;print(platform.architecture()[0]);print(platform.machine())"

7.If you want to use pip <https://pypi.org/project/pip/>_ to install PaddlePaddle, you can directly use the following command:

7. If you want to use `pip <https://pypi.org/project/pip/>`_ to install PaddlePaddle, you can use the command below directly:

(1). **CPU version** : If you only want to install CPU version, please refer to command below

Command to install CPU version is:
::

python -m pip install paddlepaddle==2.6.0 -i https://mirror.baidu.com/pypi/simple

or

python -m pip install paddlepaddle==2.6.0 -i https://pypi.tuna.tsinghua.edu.cn/simple


(2). **GPU version** : If you only want to install GPU version, please refer to command below


Note:

* You need to confirm that your GPU meets the requirements listed above

::

python -m pip install paddlepaddle-gpu==2.6.0 -i https://mirror.baidu.com/pypi/simple

or

python -m pip install paddlepaddle-gpu==2.6.0 -i https://pypi.tuna.tsinghua.edu.cn/simple
Note:

* If you want to install paddlepaddle, the version requires libstdc++.so.6 version greater than 3.4.25. To meet this requirement, you can choose to install GCC 8 or a higher GCC version, or upgrade the libstdc++ library separately. *

Please confirm that the Python where you need to install PaddlePaddle is your expected location, because your computer may have multiple Python. Depending on the environment, you may need to replace Python in all command lines in the instructions with Python 3 or specific Python path.
(1). **CPU version** : If you just want to install the CPU version, please refer to the following command for installation

8. Verify installation
The command to install the CPU version is:
::
python -m pip install --pre paddlepaddle -i https://www.paddlepaddle.org.cn/packages/nightly/cpu/

After the installation is complete, you can use `python` to enter the Python interpreter and then use `import paddle` and then `paddle.utils.run_check()` to verify that the installation was successful.
(2). **GPU version** : If you want to use the GPU version, please refer to the following command for installation

If `PaddlePaddle is installed successfully!` appears, it means the installation was successful.
The command to install the GPU cuda12.6 version is:
::
python -m pip install --pre paddlepaddle-gpu -i https://www.paddlepaddle.org.cn/packages/nightly/cu126/

The command to install the GPU cuda11.8 version is:
::
python -m pip install --pre paddlepaddle-gpu -i https://www.paddlepaddle.org.cn/packages/nightly/cu118/

9. For more information to help, please refer to:
Please make sure that the Python where you need to install PaddlePaddle is in the expected location, as your computer may have multiple Pythons. Depending on your environment, you may need to replace all python in the command line in the instructions with the specific Python path.

`install under Ubuntu <pip/linux-pip_en.html>`_
8.Verify the installation

`install under macOS <pip/macos-pip_en.html>`_
Use python to enter the python interpreter, enter import paddle, then enter paddle.utils.run_check().

`install under Windows <pip/windows-pip_en.html>`_
If PaddlePaddle is installed successfully! appears, it means you have successfully installed.

9.For more help information, please refer to:

`PIP Installation under Linux <pip/linux-pip.html>`_

The second way to install: compile and install with container
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
`PIP Installation under macOS <pip/macos-pip.html>`_

- We recommend that you use `NVIDIA PaddlePaddle Container <https://www.paddlepaddle.org.cn/documentation/docs/zh/install/install_NGC_PaddlePaddle_ch.html>`_ for your development environment installation.
- Pros
1. Lastest version of CUDA
2. Newer verison of Ubuntu OS(18.04)
3. Performance and development efficiency have been optimized by NVIDIA
`PIP Installation under Windows <pip/windows-pip.html>`_

**Second Installation Method: Using Source Code Compilation for Installation**

The third way to install: compile and install with source code
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
- If you are just using PaddlePaddle, it is recommended to use **pip** for installation.

- If you use PaddlePaddle only, we suggest you installation methods **pip** to install.
- If you need to develop PaddlePaddle, please refer to `compile from source code <compile/fromsource_en.html>`_
- If you have a need to develop PaddlePaddle, please refer to: `Compiling from Source <compile/fromsource.html>`_

.. toctree::
:hidden:
Expand All @@ -220,6 +138,5 @@ The third way to install: compile and install with source code
conda/fromconda_en.rst
docker/fromdocker_en.rst
compile/fromsource_en.rst
install_xpu_en.md
install_NGC_PaddlePaddle_en.rst
Tables_en.md